Chronic obstructive pulmonary disease (COPD) is now the third leading cause of death in the United States, surpassing stroke in 2008. The natural history of COPD is punctuated by recurrent acute exacerbations of COPD, which are associated with significant morbidity and healthcare costs. In 2010, there were about 699,000 COPD hospitalizations in the US, costing the nation $13 billion dollars in hospital bills. Furthermore, 1 in 5 hospitalizations is followed by a readmission within 30 days of discharge. To address this high-readmission issue, the Centers for Medicare and Medicaid Services (CMS) is proposing to add 30-day all-cause mortality and readmission rates for COPD as new publicly-reported quality measures. In addition, under the Affordable Care Act, CMS has been authorized to financially penalize hospitals for excessive readmissions for an initial set of three conditions (acute myocardial infarction, heart failure, and pneumonia) since October 1, 2012. Beginning in fiscal year 2015, the list will expand to include COPD. However, there is a dearth of research on COPD readmission, such as lack of a validated risk-adjustment model and scant information on best treatment strategies to reduce COPD readmissions. We propose to compile and analyze national, longitudinal data from four sources (the Healthcare Cost and Utilization Project [HCUP], the Medicare Claims/Part D drug data, the Medicare Current Beneficiary Survey [MCBS], and the National Health and Nutrition Examination Survey [NHANES]-Medicare linked data) to address these knowledge gaps. The Specific Aims are: (1) To examine national trends in 30-day readmission rates (all-cause and cause-specific) after hospitalization for COPD during 2006-2012, both overall and by race/ethnicity, US regions, number of comorbid conditions, and AHRQ- defined priority populations (e.g., women and the elderly); (2) To develop and validate three risk-adjustment models for 30-day all-cause COPD readmission: models based on administrative claims alone (Medicare), survey-enhanced claims (MCBS), and clinically-enriched claims (NHANES-Medicare); (3) To compare the effectiveness of initiation of guideline-recommended interventions in reducing all-cause and COPD-related rehospitalizations at 30 days and 1 year. The interventions will include 5 broad strategies encompassing 8 comparisons: smoking cessation (e.g., varenicline vs. bupropion), vaccines (e.g., influenza and pneumococcal), bronchodilators/combination therapies, pulmonary rehabilitation, and follow-up with a primary care provider. The proposed study is highly significant because it addresses both clinical and methodological aspects of CER on COPD readmission, an understudied but very important topic. The study will generate novel results including epidemiologic analysis to identify subpopulations at higher risk for readmissions, three new risk-adjustment/prediction models, and information on best treatment options in reducing COPD readmissions. Accordingly, we expect that the results will help reduce costly readmissions and have important implications for patients, family members, clinicians, health insurers, hospital administrators, and policymakers.